Predictive Method for Controlling a Blood Processing System

• US 20100025336A1
• Filed: 09/17/2009
• Published: 02/04/2010
• Est. Priority Date: 07/02/2003
• Status: Active Grant

• Current Assignee: Terumo BCT, Inc. (Terumo Corporation)
• Original Assignee: Caridianbct Biotechnologies LLC (Baxter International Inc.)
• Inventors: KOLENBRANDER, Jeremy P., LADTKOW, James R., STEWARD, Jeffrey A., CARTER, Lee F., SCIBONA, Joseph A., FLETCHER, Chris
• Granted Patent: US 7,906,771 B2
• Time in Patent Office: Days
• US Class Current: 210/740
• CPC Class Codes: A61M 1/3693 using separation based on different densities of components, e.g. centrifuging
• CPC Class Codes: B04B 13/00 Control arrangements specially designed for centrifuges; Programme control of centrifuges control arrangements for feed, charge, or discharge B04B11/00
• CPC Class Codes: B04B 2013/006 Interface detection or monitoring of separated components
• CPC Class Codes: G01N 15/042 by centrifuging and investigating centrifugates centrifuges per se B04B
• CPC Class Codes: G01N 15/05 in blood
• CPC Class Codes: G01N 2015/047 by static multidetectors

Abstract

The invention relates generally to methods of monitoring and controlling the processing of blood and blood samples, particularly the separation of blood and blood samples into its components. In one aspect, the invention relates to optical methods for measuring two-dimensional distributions of transmitted light intensities, scattered light intensities or both from a separation chamber of a density centrifuge. The method may include performing first and second measurements of an operating condition; analyzing the first and second measurements using a predictive data analysis algorithm; comparing the predicted operating condition to a desired operating condition; and adjusting at least one setting.

Claims

1. A method for controlling a density centrifuge blood processing system for separating fluid components, said method comprising the steps of:

performing a first measurement of an operating condition of said blood processing device corresponding to a first time;

performing a second measurement of said operating condition of said blood processing device corresponding to a second time;

analyzing said first and second measurements of said operating condition using a predictive data analysis algorithm, wherein operation of said predictive data analysis algorithm generates a predicted operating condition of said blood processing device corresponding to a selected future time;

comparing the predicted operating condition of said blood processing device corresponding to the selected future time to a desired operating condition; and

adjusting at least one setting of said blood processing device based on said comparison of the predicted operating condition of said blood processing device at said future time and the desired operating condition, thereby controlling said blood processing device.

2. The method of claim 1 wherein said first and second measurements comprise measurement of a first and a second position of a phase boundary between optically differentiable fluid components in the density centrifuge blood processing system.

3. The method of claim 2 further comprising tracking cells flowing out a selected extraction port.

4. The method of claim 3 further comprising concurrently measuring a red blood cell phase boundary and tracking collected blood component concentration in the extraction port.

5. The method of claim 3 further comprising concurrently measuring a buffy coat layer and tracking a presence of red blood cells in the extraction port.

6. The method of claim 1 wherein said first and second measurements comprise measurement of a first and a second composition of fluid components in an extraction port of the density centrifuge blood processing system.

7. The method of claim 1 further comprising

analyzing an object data list every time a new image is acquired and

analyzing the object data list as pair of chronologically ordered frames for comparative analysis.

8. The method of claim 7 further comprising

acquiring and retaining a previous frame containing a number of chronologically ordered image data objects;

and

acquiring a current frame containing a most recently acquired image data object, and a specified number chronologically ordered data objects that immediately preceded the most recently acquired data object,

wherein the data objects in the previous frame match the data objects in the current frame starting with the image data object sequenced immediately before the oldest image data object in the current frame.

9. The method of claim 8 wherein said predictive data analysis algorithm compare and correlate a plurality of parameters from the two frames to derive positional, directional, characteristic, and associated rates of change information relating to selected extracted image data information.

10. The method of claim 9 wherein discrete magnitudes of changes in a plurality of parameters as a function of corresponding discrete time intervals are used to derive velocity and acceleration information for selected parameters.

11. The method of claim 1 further comprising

ordering a series of image frames corresponding to light intensities to be collected;

specifying a first number of frames for measuring a cell interface position and an optical density of fluid in an extraction port; and

specifying a second number of frames, less than said first number, for image analysis to monitor image data relating to the quality of the images being collected, and thus relating to the reliability of the measurements collected in the other frames.

12. The method of claim 11 further comprising splitting data from said first number of frames into a first data stream and data from said second number of frames into a second data stream.

13. The method of claim 12 wherein said second data stream is used by an automated process control system to determine reliablity of measurements and to adjust device parameters to improve image quality.